JP2003345389A - Voice recognition device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice recognition device which enables a driver and passengers other than the driver to easily perform voice input operation. <P>SOLUTION: The voice recognition device which is mounted on a vehicle and recognizes a voice spoken by a passenger to output operation information to various apparatus provided in the vehicle is provided with a driver's seat-side speech switch 2, an assistant driver's seat-side speech switch 3, and a rear seat-side speech switch 4 which permit voice input operation by operation of passengers. The driver's seat-side speech switch 2 is provided in the vicinity of a driver's seat, and the assistant driver's seat-side speech switch 3 is provided in the vicinity of an assistant driver's seat, and the rear seat-side speech switch 4 is provided in the vicinity of a rear seat. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice recognition device mounted on a vehicle.

[0002]

2. Description of the Related Art Conventionally, for example, JP-A-2001-29
As shown in Japanese Patent No. 6891, as a voice recognition device mounted on a vehicle, when a voice is input to a voice input unit (microphone or the like), an input angle (directional characteristic) of the voice is measured,
There is known a voice recognition device that determines which seat an occupant uttered the voice based on the measured input angle.

[0003]

However, the voice recognition device has the following problems.

That is, since the inside of the vehicle is relatively small, the input angles of the voices uttered by the passengers sitting in different seats may be similar to each other, so the passengers in any seat uttered the voice. It was not easy to accurately determine whether or not the input angle of the voice.

Further, since the input angle of the voice changes depending on the sitting posture of the occupant who speaks the voice, in this respect also, it is possible to determine which seat the occupant uttered the voice. It was not easy to make an accurate judgment based on the input angle.

Further, in the voice recognition device, in order to start the voice input operation, a portion near the driver's seat (for example, a steering column, an instrument panel, etc.)
I had to operate the speech switch provided in.

For this reason, it is not easy for the passenger sitting in the passenger seat or the rear seat to press the utterance switch, so that the voice input operation cannot be easily performed.

For this reason, the occupant sitting in the passenger seat or the rear seat first asks the vehicle driver to press the utterance switch, and then either speaks voice information by himself or asks the vehicle driver to speak. The voice input operation was performed by having the voice information uttered.

Therefore, the occupant sitting in the passenger seat or the rear seat needs the driver to press the utterance switch every time he / she wants to perform the voice input operation, which is troublesome for the voice input operation. .

Some voice recognition devices limit voice input operation while the vehicle is running.

Therefore, while the vehicle is traveling, the voice input operation by the passenger sitting in the passenger seat or the rear seat is also restricted.

The present invention has been made to solve such conventional problems, and its main purpose is to allow a driver and an occupant other than the driver to easily perform a voice input operation. It is to provide a voice recognition device that can do it.

[0013]

In order to achieve the above object, the invention according to claim 1 of the present application is mounted on a vehicle, recognizes a voice uttered by an occupant, and provides various devices to the vehicle. A voice recognition device that outputs operation information is provided with a plurality of voice input starting means for enabling voice input operation by an operation by an occupant, and the voice input starting means is in the vicinity of the driver's seat and seats other than the driver's seat. It is characterized in that it is provided in the vicinity of at least one of the seats.

According to a second aspect of the invention, in the voice recognition apparatus according to the first aspect, the output of the operation information to the various devices is restricted according to the operation input of each of the voice input start means. To do.

The invention according to claim 3 is the invention according to claim 1 or 2.
In the voice recognition device described above, the output of the operation information to the various devices is limited according to the traveling state of the vehicle.

[0016]

According to the first aspect of the present invention, the driver operates the voice input starting means provided near the driver's seat (that is, the voice input starting means that is easy for the driver to operate). Voice input operation can be performed.

The occupant sitting in at least one of the seats other than the driver's seat (hereinafter referred to as "specific seat") is a voice input start means (provided near the specific seat). That is, the voice input operation can be performed by operating the voice input start means) which is easy for the occupant to operate.

Therefore, the driver and passengers other than the driver can easily perform the voice input operation.

According to the second aspect of the invention, in addition to the effect of the first aspect of the invention, the output of the operation information to various devices is restricted according to the operation input of each voice input start means, so the operation input is performed. Accordingly, it is possible to prevent various devices from performing processing corresponding to the content of the voice spoken by the occupant. In other words, only the voice input operation according to the operation input can be made possible.

Therefore, for example, in a vehicle equipped with a driver side air conditioner for adjusting the temperature around the driver's seat as various devices, only the voice input start means provided near the driver's seat is operated and input, and the temperature information is displayed. (For example, "25
When the temperature information "degree") is uttered, the output of the operation information can be limited so that only the air conditioner on the driver's side operates.

In this case, even if an occupant sitting in a specific seat operates the voice input start means provided in the vicinity of the specific seat and speaks temperature information, the air conditioner on the driver side does not operate.

According to the third aspect of the present invention, in addition to the effect of the first or second aspect of the invention, the output of the operation information to various devices is restricted according to the running state of the vehicle. Thus, it is possible to prevent various devices from performing processing corresponding to the content of the voice spoken by the occupant. In other words, it is possible to enable the voice input operation according to the traveling state.

Thus, for example, when the vehicle is traveling and only the voice input start means provided near the driver's seat is operated and input, the operation information to the display device included in various devices is displayed. If the output of is limited, even if the driver speaks "TV" after the driver has operated the voice input start means provided near the driver's seat when the vehicle is traveling, It is possible not to display the television image on the display device.

Thus, it is possible to prevent the driver from operating the television while the vehicle is traveling.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

First, the configuration of the speech recognition apparatus 1 according to the present invention will be described with reference to FIGS. 1 and 2. here,
FIG. 1 is a block diagram showing the configuration of the voice recognition device 1, and FIG. 2 is a schematic diagram showing the arrangement of a driver side speech switch 2 (described later) and the like included in the voice recognition device 1.

The voice recognition apparatus 1 is a vehicle 100 shown in FIG.
As shown in FIG. 1, the driver's seat side utterance switch (voice input start means) 2, the passenger seat side utterance switch (voice input start means) 3, and the rear seat side utterance switch (voice input start) 4), a plurality of standby dictionary databases 5 (only one standby dictionary database 5 is shown in FIG. 1), a switch determination unit 6 for determining which utterance switch has been operated, and the vehicle 100. A traveling detection unit 7 for detecting the traveling state of the vehicle, a microphone 8 used for a voice input operation, a voice recognition unit 9 for performing voice recognition and the like based on the result of the determination by the switch determination unit 6, and a voice recognition unit. An operation information output unit 10 that outputs operation information to various devices provided in the vehicle 100 according to the result of the voice recognition by the vehicle 9.

When the driver's seat side utterance switch 2 is pressed (operation input) by an occupant (as will be described later, the occupant becomes the driver of the vehicle 100), it is indicated that the driver has requested a voice input operation. The voice input operation request information is output to the switch determination unit 6.

Further, since the driver side utterance switch 2 is provided on the steering wheel 101 (that is, provided near the driver's seat) as shown in FIG. It is easy to press the seat side speech switch 2, but it is difficult for other passengers to press it.

When the passenger side utterance switch 3 shown in FIG. 1 is pressed (operation input) by an occupant (as will be described later, the occupant becomes an occupant sitting in the passenger seat), the occupant switch sits in the passenger seat. The voice input operation request information indicating that the voice input operation is requested by the occupant who is present is output to the switch determination unit 6.

The passenger side utterance switch 3 is provided near the passenger side door grip 102 as shown in FIG. 2 (that is, near the passenger seat (a seat other than the driver's seat)). Therefore, it is easy for the passenger sitting in the passenger seat to press the passenger side utterance switch 3, but it is difficult for other passengers to press it.

The rear seat side speech switch 4 shown in FIG.
When the occupant (the occupant becomes a occupant sitting in the rear seat (a seat other than the driver's seat) as described later) is pressed (operation input), voice input operation is requested by the occupant sitting in the rear seat. The voice input operation request information to the effect that it has been output to the switch determination unit 6.

The rear seat side speech switch 4 (not shown in FIG. 2) is the console box 10 shown in FIG.
3 is provided at the rear end (that is, provided in the vicinity of the rear seat), it is easy for an occupant sitting in the rear seat to press the rear seat side utterance switch 4.
It is difficult for other occupants to press.

The standby dictionary database 5 ... Stores words that may be uttered by an occupant, but these standby dictionary databases 5 ... Store different words.

Then, the standby dictionary database 5
..., the driver-only standby dictionary database 5a,
A passenger-side standby dictionary database 5b and a rear-seat standby dictionary database 5c are included.

Here, the driver-side standby dictionary database 5a stores words that may be uttered by the driver, and the passenger-side standby dictionary database 5a.
b stores the words that may be uttered by the passenger sitting in the passenger seat, and the rear seat dedicated standby dictionary database 5c stores the words that may be uttered by the passenger sitting in the rear seat. To do.

The switch determination section 6 presses which utterance switch is pressed based on the voice input operation request information given from the driver seat utterance switch 2, the passenger seat utterance switch 3, and the rear seat utterance switch 4. To determine if.

Then, based on the determination result, a predetermined standby dictionary database 5 is selected from the standby dictionary databases 5 ... And the selected standby dictionary database 5 is set in a state accessible by the voice recognition unit 9. . Also, the determination result information regarding the determination result is created and output to the voice recognition unit 9. The selection procedure will be described later.

The traveling detection unit 7 detects the traveling state of the vehicle and creates detection result information.

The microphone 8 is, as shown in FIG.
0 is attached to the rearview mirror 105 and is in a voice input standby state by the process of the voice recognition unit 9 (described later), the voice uttered by the occupant is captured, and the captured voice is A / The D-converted voice information is output, and the voice information is output to the voice recognition unit 9.

The voice recognition unit 9 sets the microphone 8 to a state in which the occupant can perform a voice input operation, that is, a voice input standby state, based on the determination result information provided from the switch determination unit 6.

Therefore, the occupant can perform a voice input operation by depressing any one of the driver side speech switch 2, the passenger side speech switch 3, and the rear seat side speech switch 4.

When voice information is given from the microphone 8, the voice information is analyzed to recognize the contents, and the matching process described later is performed on the contents of the voice information.

Then, predetermined operation result information is created based on the result of the matching process, and the operation information output unit 10 is operated.
Output to. The details will be described later.

The operation information output unit 10 acquires the detection result information from the traveling detection unit 7, and outputs the operation information based on the detection result information and the recognition result information given from the voice recognition unit 9. Decide whether or not.

When it is decided to output, predetermined operation information is created, and the predetermined operation information is output to a predetermined device among various devices provided in the vehicle 100 shown in FIG. The details will be described later.

The various devices include, for example, a display device having the front display 104 shown in FIG. 2 and a rear display not shown, and a speaker not shown.
It includes a driver side air conditioner, a passenger side air conditioner, an electronic mail transmitting / receiving device, a television receiving device, and a route guiding device.

Here, the front display 104 is shown in FIG.
As shown in FIG. 5, it is provided in front of the driver's seat and the passenger seat, and is mainly used by the driver and the passenger sitting in the passenger seat. The rear display is provided, for example, at the rear end of the console box 103 and is mainly used by an occupant sitting in the rear seat.

The driver side air conditioner can adjust the temperature around the driver's seat, and the passenger seat side air conditioner can adjust the temperature around the passenger seat.

Further, the electronic mail transmission / reception device performs a process that enables the passenger to create a mail text by an operation described later by the passenger. The mail text is created by a voice input operation using the microphone 8.

Next, the procedure of processing by the voice recognition device 1 will be described with reference to FIGS. Here, FIG.
3 is a flowchart showing a procedure of processing by the voice recognition device 1.

In step S1 shown in FIG. 3, the occupant who wants to perform a voice input operation selects one of the driver side speech switch 2, the passenger side speech switch 3 and the rear seat side speech switch 4 shown in FIG. Press the utterance switch that is most easily pressed by the passenger.

That is, when the occupant is a driver,
When the driver-side utterance switch 2 is pressed and the occupant is a passenger sitting in the passenger seat, when the passenger-side utterance switch 3 is pressed and the occupant is a passenger sitting in the rear seat, Pushes the speech switch 4 on the rear seat side.

Next, the pressed speech switch is
The voice input operation request information is output to the switch determination unit 6.

In step S2, the switch determination section 6
Which utterance switch is pressed is determined based on the voice input operation request information.

Next, when it is determined that the driver side utterance switch 2 has been pressed, in step S3a, the driver seat dedicated standby dictionary database 5a is selected from the standby dictionary databases 5 ... The dictionary database 5a is set to be accessible by the voice recognition unit 9.

On the other hand, if it is determined that the passenger seat side speech switch 3 has been pressed, in step S3b, the passenger seat dedicated standby dictionary database 5b is selected from the standby dictionary databases 5 ... The dictionary database 5b is set to be accessible by the voice recognition unit 9.

On the other hand, when it is determined that the rear seat side utterance switch 4 is pressed, the rear seat dedicated standby database 5c is selected from the standby dictionary databases 5 ... In step S3c, and the rear seat dedicated standby is selected. The dictionary database 5c is set to be accessible by the voice recognition unit 9.

Next, the judgment result information on the judgment result is created and output to the voice recognition section 9.

Then, in step S4, the voice recognition unit 9 shown in FIG. 1 sets the microphone 8 in the voice input waiting state based on the determination result information provided from the switch determination unit 6.

In step S5, when the occupant utters the contents of the desired processing (for example, utters "TV" if he / she wants to watch TV), the microphone 8 captures the uttered sound. Then, the captured voice is A / D converted into voice information, and the voice information is output to the voice recognition unit 9.

On the other hand, the running detection unit 7 is the vehicle 1 shown in FIG.
The running state of 00 is detected to create detection result information.

In step S6, the voice recognition unit 9 analyzes the voice information given from the microphone 8 to recognize the contents,
The following matching processing is performed on the content of the voice information.

That is, the standby dictionary database 5 made accessible by the processing by the switch determination unit 6 is accessed, and the degree of matching with the content of the voice information is measured for each word stored in the standby dictionary database 5. , If there is a word whose matching degree is equal to or higher than a predetermined value (the predetermined value is determined in advance), then the word (if there are a plurality of words whose matching degree is equal to or higher than the predetermined value, Get the word with the highest degree of matching).

For example, when the content of the voice information is “TV”, the word “TV” having the highest matching degree is acquired from the standby dictionary database 5.

Next, the first recognition result information including the word information regarding the acquired word and the utterance switch information regarding the type of the uttered switch that has been pressed is created and output to the operation information output unit 10.

On the other hand, when there is no word having a matching degree of a predetermined value or more, second recognition result information indicating that there is no word having a matching degree of a predetermined value or more is created and output to the operation information output unit 10. To do.

Next, the operation information output unit 10 performs the following processing based on the recognition result information given from the voice recognition unit 9.

That is, when the second recognition result information is given, the operation information requesting the presentation of the information prompting the voice input operation again is output to the speaker and the display device.

As a result, the speaker outputs voice information for prompting another voice input operation (for example, voice information "Please speak again"), and the display device displays image information for prompting another voice input operation. indicate.

Here, the standby dictionary database 5 ...
Memorize words different from each other, therefore, even if the voice information is the same, depending on the type of the standby dictionary database 5 that can be accessed, there is a word with a matching degree of a predetermined value or more, and a case with no word. There is.

The type of the standby dictionary database 5 that can be accessed is determined based on the type of the speech switch depressed by the occupant. That is, it is determined according to the operation input of the speech switch.

Therefore, even if the voice information is the same,
Depending on the type of speech switch pressed by the occupant,
The operation information according to the content of the voice information may not be output to various devices. That is, the voice recognition device 1 may limit the output of the operation information by the operation information output unit 10.

On the other hand, when the first recognition result information is given, first, based on the first recognition result information, the content of the word acquired by the voice recognition unit 9 and the operator's depression. By recognizing the type of speech switch, the content of the process requested by the operator is determined.

For example, if the driver is the driver side speech switch 2
When pressing and uttering "TV" (in this case, the content of the word recognized by the operation information output unit 10 is "TV", and the type of utterance switch is the driver side utterance switch 2), The operation information output unit 10 determines that the content of the process requested by the operator is “to display a television image on the front display 104 shown in FIG. 2”.

Next, the detection result information is acquired from the traveling detection unit 7 shown in FIG. 1, and based on the detection result information shown in FIG.
The running state of the vehicle 100 shown in is determined.

As a result, when it is determined that the vehicle 100 is stopped, the operation information corresponding to the processing requested by the occupant is created and output to a predetermined device.

As a result, the device to which the operation information is given performs the processing requested by the occupant.

On the other hand, when it is determined that the vehicle 100 is traveling, it is determined whether or not the process requested by the occupant is a process not permitted during driving.

Here, as the processing not permitted during driving, there is, for example, processing for displaying a television image on the front display 104.

As a result, when it is determined that the process requested by the occupant is a process not permitted during driving, the operation information is not output.

On the other hand, when it is determined that the process requested by the occupant is not a disallowed process during operation, operation information corresponding to the process requested by the occupant is created and output to a predetermined device. As a result, the device given the operation information performs the processing requested by the occupant.

Therefore, the voice recognition device 1 limits the output of the operation information by the operation information output unit 10 according to the traveling state of the vehicle.

Next, an example of processing by the voice recognition device 1 will be described with reference to FIGS.

For example, after the occupant sitting in the driver's seat (that is, the driver) presses the driver-side utterance switch 2, "25
"(In this case, the process requested by the occupant is a process of adjusting only the temperature around the driver's seat to" 25 degrees "), the voice recognition apparatus 1 performs the above-described step S1.
-Process of step S3a and step S4-step S
By the process of 6, the operation information for requesting that the temperature around the driver's seat be "25 degrees" is output to the air conditioner on the driver's seat side.

As a result, the driver side air conditioner adjusts only the temperature around the driver's seat to 25 degrees.

Similarly, when the passenger sitting in the passenger seat utters "25 degrees" after pressing the passenger side utterance switch 3, (in this case, the processing requested by the passenger is the temperature around the passenger seat). It becomes the process of adjusting only to "25 degrees"),
The voice recognition device 1 uses the steps S1 to S described above.
Through the processing of 3b and the processing of steps S4 to S6, operation information requesting that the temperature around the passenger seat be set to "25 degrees" is output to the passenger side air conditioner.

As a result, the passenger side air conditioner adjusts only the temperature around the passenger seat to 25 degrees.

Further, when the vehicle is running, after the driver presses the driver-side utterance switch 2, "TV"
(In this case, the process requested by the occupant is a process of displaying a television image on the front display 104, but the process is a process that is not permitted during driving.) The operation information is not output by the processes of steps S1 to S3a and steps S4 to S6 described above.

As a result, the front display 104 is
Since the television image is not displayed, it is possible to prevent the driver from operating the television when the vehicle 100 is traveling.

When the vehicle 100 is running,
Even when the passenger sitting in the passenger seat utters "TV" after pressing the passenger side utterance switch 3, the voice recognition device 1 performs the above-described steps S1 to S3b and steps S4 to S6. The operation information is not output by the processing of. In this case as well, the processing requested by the occupant is processing for displaying the television image on the front display 104, that is, processing that is not permitted during driving.

On the other hand, when the occupant sitting in the rear seat utters "TV" after pressing the rear seat side utterance switch 4 (in this case, the processing requested by the occupant causes the TV image to be displayed on the rear display). Although it is a process of displaying, the process is not a process of prohibition during driving), the voice recognition device 1 performs the process of steps S1 to S3c and the process of steps S4 to S6 described above to display the television image. The operation information requesting display on the rear display is output to the television receiving device.

As a result, the television receiving device displays the television image on the rear display.

When the vehicle 100 is running,
When the driver utters "mail text input" after pressing the driver side utterance switch 2 (in this case, the process requested by the occupant is a process that enables the driver to create a mail text, The process is a non-permission process during driving), and the voice recognition device 1 does not output the operation information due to the processes of steps S1 to S3a and steps S4 to S6 described above.

As a result, the mail transmitting / receiving device does not perform the processing that enables the driver to create a mail text, and thus prevents the driver from creating a mail text while the vehicle 100 is traveling. You can

On the other hand, when the vehicle 100 is running,
When the passenger sitting in the passenger seat presses the passenger side utterance switch 3 and then utters “mail text input” (in this case, the processing requested by the passenger is the mail text by the passenger sitting in the passenger seat). However, the process is not a process of disallowing during driving), and the voice recognition device 1 uses the steps S1 to S3 described above.
Through the processing of b and the processing of steps S4 to S6, the operation information requesting that the passenger sitting in the passenger seat be able to create the mail text is output to the mail transmitting / receiving device.

As a result, the mail transmitting / receiving device performs a process that enables the passenger sitting in the passenger seat to create a mail text, so that the passenger sitting in the passenger seat can perform a voice input operation using the microphone 8. You can create an email text.

As described above, according to the present embodiment, the voice recognition device 1 includes the driver side speech switch 2 provided near the driver's seat and the passenger side speech switch 3 provided near the passenger seat. And the rear seat side speech switch 4 provided in the vicinity of the rear seat, the driver operates the driver seat side speech switch 2 (that is, the speech switch that is most easily pressed by the driver). A voice input operation can be performed by pressing the button.

Similarly, the passenger sitting in the passenger seat can perform a voice input operation by pressing the passenger seat side utterance switch 3 (that is, the utterance switch that is most easily pressed by the passenger). Further, the occupant sitting in the rear seat can perform voice input operation by pressing the rear seat side utterance switch 4 (that is, the utterance switch that is most easily pressed by the occupant).

Therefore, the driver and the occupants other than the driver can easily perform the voice input operation.

Further, since the output of the operation information to various devices is restricted in accordance with the operation input of each utterance switch, the processing corresponding to the contents of the voice uttered by the occupant is performed in various devices according to the operation input. You can choose not to do it. In other words, it is possible to enable the voice input operation according to the operation input.

Further, since the output of the operation information to various devices is restricted according to the running state of the vehicle 100, the processing corresponding to the content of the voice uttered by the occupant is performed on the various devices according to the running state. You can prevent it. In other words, it is possible to enable the voice input operation according to the traveling state.

In this embodiment, the driver's seat side speech switch 2 is provided on the steering wheel 101, but it may be provided anywhere near the driver's seat.

Although the passenger seat side speech switch 3 is provided near the passenger seat side door grip 102, it may be provided anywhere near the passenger seat.

Further, although the rear seat side speech switch 4 is provided at the rear end of the console box 103, it may be provided at any position in the vicinity of the rear seat.

When the vehicle 100 is running, and the processing requested by the occupant is driving and is not permitted, the operation information is not output to various devices. In this case, The operation information indicating that the processing requested by the occupant cannot be performed may be output to the speaker and the display device.

As a result, the speaker gives voice information indicating that the processing requested by the occupant cannot be performed (for example, "the processing cannot be performed while running").
Voice information), and the display device can display image information indicating that the processing requested by the occupant cannot be performed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a voice recognition device according to the present invention.

FIG. 2 is a schematic view showing a position where a driver's seat side speech switch and the like included in the voice recognition device are provided.

FIG. 3 is a flowchart showing a procedure of processing by a voice recognition device.

[Explanation of symbols]

1 Speech recognition device 2 Driver side utterance switch (voice input start means) 3 Speech switch on the passenger side (voice input start means) 4 Rear seat side speech switch (voice input start means) 5 Standby dictionary database 6 Switch determination unit 7 Running detector 8 microphone 9 Speech recognition unit 10 Operation information output section

─────────────────────────────────────────────────── ─── Continued Front Page (51) Int.Cl. ⁷ Identification Code FI Theme Coat (Reference) G10L 15/24 G10L 3/00 521V

Claims (3)

[Claims] 1. A voice recognition device mounted on a vehicle for recognizing a voice uttered by an occupant and outputting operation information to various devices provided in the vehicle, the voice input operation being enabled by an operation by the occupant. A plurality of voice input start means, the voice input start means is in the vicinity of the driver's seat and in the vicinity of at least one seat other than the driver's seat,
A voice recognition device, which is provided in each. 2. The voice recognition device according to claim 1, wherein the output of the operation information to the various devices is restricted according to the operation input of each of the voice input start means. 3. The voice recognition device according to claim 1, wherein output of operation information to the various devices is restricted according to a running state of the vehicle.